List your company
ResearchFormerly WizWand

Single-hop Question Answering on Factrecall en

31.36ScoreLM-infinite
1.12728.976116.82524.6739Jun 1, 2025
Updated 8mo ago

Evaluation results

MethodLinks
LM-infiniteBackbone=Mistral-7B-Instruct2025.0631.36
DCSBackbone=Llama-3-8B-Instruct2025.0629.89
StreamingBackbone=Mistral-7B-Instruct2025.0629.64
DCSBackbone=Vicuna-7B2025.0627.26
InfLLMBackbone=Mistral-7B-Instruct2025.0624.64
InfLLMBackbone=Llama-3-8B-Instruct2025.0619.22
InfLLMBackbone=Vicuna-7B2025.0616.65
Llama-3-8B-InstructBackbone=Llama-3-8B-Instruct, Context Management=Baseline2025.0615.5
StreamingBackbone=Llama-3-8B-Instruct2025.0612.36
LM-infiniteBackbone=Llama-3-8B-Instruct2025.0612.16
MOICEBackbone=Vicuna-7B2025.068.27
Vicuna-7BBackbone=Vicuna-7B, Context Management=Baseline2025.066.81
DCSBackbone=Mistral-7B-Instruct2025.066.64
LM-InfiniteBackbone=Vicuna-7B2025.063.3
StreamingBackbone=Vicuna-7B2025.062.74
MOICEBackbone=Mistral-7B-Instruct2025.062.64
Mistral-7B-InstructBackbone=Mistral-7B-Instruct, Context Management=Baseline2025.062.29